In his remarks to lawmakers Tuesday, Federal Reserve Chair Jerome Powell talked about a reversal in the softening of many of the metrics he’s been watching: employment, consumer spending, inflation. And he spent a quick second talking about how the unusually warm weather in January may have messed with seasonally adjusted data, making the numbers seem excessively strong.

“Seasonally adjusted” is a term we throw around when we spit out facts and figures about the economy. But it’s an important enough idea that we wanted to slow down and explain exactly what it means in layman’s terms.

“There are people who do this for a living, and it really, really, really matters,” said economist Justin Wolfers at the University of Michigan. That’s because seasonal adjustments allow us to factor in wild swings.

Say we own an ice cream that sells a lot during summer and a little during winter. We know not to freak out when it starts snowing and sales dip, and we’re better able to see when something unusual happens — like a spike.

“We say, ‘Well, that must mean there’s something good going on with ice cream sales,'” Wolfers said.

Turns out our new brownie ice cream is drawing customers.

Now, the government does this too because our economy is always fluctuating. Like, if we didn’t use seasonally adjusted numbers, it would look like we’re heading into a recession every January after the holidays when retail sales tank.

“But every once in a while, they don’t quite happen to exactly the same degree, and that causes some confusion about what the data in a particular month or quarter actually mean,” said Jeffrey Miron, vice president of research at the Cato Institute.

The confusion is figuring out the outliers — the brownie ice creams. Sometimes they’re obvious. Think job numbers tanking because of COVID lockdowns or oil prices spiking because of the war in Ukraine. But January’s outlier? Maybe the unusually warm weather coaxed people to go outside … and fueled inflation. 

“It’s very hard to explain that all with weather,” said Jonathan Wright, an economist at Johns Hopkins. 

It’s tough to trust data when the past three Januarys have been so weird, he said. Though seasonal adjustments are based on several years of data, we’re not in the same economy as 2019.

So there’s a risk that “you can either overstate or understate the true strength of the economy,” Wright added.